FEDERATED HEALTH REGISTERS

A Literature Review

Liliana Leite

, Alice Miranda

, Carla Esteves

and Luis Antunes

Faculty of Medicine, University of Porto, Alameda Professor Hernâni Monteiro, 4200-319 Porto, Portugal

Faculty of Science, University of Porto, Rua do Campo Alegre s/n, 4169- 007 Porto, Portugal

Keywords: Federated health registers, Electronic health record, Message-based integration, Federated architecture.

Abstract: Electronic Health Records (EHR) are a collection of all individuals health data, in an electronic form,

generated during relevant interactions with the healthcare system. The federated database systems provides

intercommunication between different and autonomous data units, which enables sharing data. This paper

aims to be a state-of-art on the requirements to take into account when developing a federated system. For

that, we did a literature review on PubMed, ISI Web of Knowledge, Scopus and Google Scholar. Federated

systems must ensure interoperability using open standards; guarantee the system value through high quality

services; have modular architecture to allow developments, maintenance and evolutions, and finally, enable

the “monotonic” systems with an incremental evolution. When developing these systems we have to avoid

semantic, functional and instance conflicts to ensure the correct functionality. We concluded that federated

systems are a good option in health's domain. They allow a high volume of data storage in healthcare that

can be accessed in any place, at any time, by health professionals. Thus, we believe that federated systems

are a tool to improve the quality and efficiency of health care.

1 INTRODUCTION

With the growth of healthcare volume data at

multiples locations for a wide of users, the need for

integrate the information emerge and the concept of

Electronic Health Records (EHR) as a central

solution appears.

EHR is a collection of all individual's lifetime

health data in an electronic form produced during

relevant interactions with the healthcare system

(Tsiknakis et al., 2002). The EHR allows electronic

documentation of current and historical health like

tests, referrals and medical treatments, as well,

enable practitioners to order tests and medications

electronically. The improvement of communication

between physicians and patients is one of the

potentiality of EHRs. They can make data more

readily available, at a corporate, regional, national or

even international level. Population mobility is

increasing, and the construction of an EHR at a high

level could be an improvement for all, specially

those people.

Federation system is a process that allows

sharing information provided by different and

independent units. In health environment, the

federated approach should be capable of providing

uniform ways for accessing authentic, physician-

generated, patient record information that is

physically located in different clinical information

systems (Katehakis et al., 2001). A federated

database system is constituted by a multiple and

autonomous databases systems integrated

transparently into a single federated database. These

databases are interconnected via a computer network

and are geographically decentralize. It is a virtual

database, fully integrated, logical composite of all

constituent databases in a federated database system

(Muilu et al., 2007).

Through a literature review, we aim to study

federated electronic health records to understand this

way of sharing and archiving information.

1.1 Background

The development of federated EHR brings some

benefits (Tsiknakis et al., 2002) such as vital health

information available at any time; patient's relevant

medical history be accessed by health practitioners

what provides more efficient and effective treatment

and more quality-time with the patient; reduction of

158

Leite L., Miranda A., Esteves C. and Antunes L..

FEDERATED HEALTH REGISTERS - A Literature Review.

DOI: 10.5220/0003732401580163

In Proceedings of the International Conference on Health Informatics (HEALTHINF-2012), pages 158-163

ISBN: 978-989-8425-88-1

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

costs and redundant procedures by accessing

information on previous medical or lab

examinations; improvement of care quality based on

an enhanced ability of health planners and

administrators to develop relevant health care

policies for the future; great empowerment of

individuals by giving them access to their own

personal health records, given them the opportunity

to exercise greater control over their own health.

Sharing medical records has the interoperability

between systems as a prerequisite and consists in a

technological chalenge (Katehakisv et al., 2001).

The implementation of such system requires

agreement between the various equipment suppliers

and will certainly increase the costs. So, the

operability of systems and services based on

standards is a key point in order to achieve the

networks’s integration of medical care (Katehakis et

al., 2001).

Modern systems are highly distributed and

heterogeneous. Interconnect data from different

sources optimize resources and enable services to

exchange information and use it in different systems

in an easier way; ensure the understanding and

preservation of the context and meaning of

information exchanged and allows coexistence of

different systems, without forcing the conversion to

a single format.

The aspects referred bellow, leads to the concept

of "interoperability" defined by Healthcare

Information Management Systems Society (HIMSS)

briefly as follows (Portuguese Health Ministry,

2009):

"Interoperability is the ability of information systems

in health work together, either within organizations

or across organizational boundaries in support of an

effective health care to individuals and the

community."

Interoperability can be materialized at various levels

like technical and semantic.

Technical interoperability ensures the integration

of different systems at the technical level,

infrastructure, media, transportation, storage and

data representation.

Semantic interoperability enables the encoding,

transmission and use of information relating to

health services among the various stakeholders,

ensuring the understanding of information, either by

their own systems, or by users.

Interoperability can be achieved by messages or

by a more advanced approach based on federated

autonomous systems (Katehakis et al., 2001). The

federated approach is used primarily to provide a

virtual view of the Integrated Electronic Health

Record (I-EHR) without replicating unnecessary

information while on the message approach there is

data redundancy.

2 METHODS

A systematic review, between 27th December 2010

and 3rd January 2011, was performed on PubMed,

ISI Web of Knowledge and Scopus. The first step

was defining our keywords: federated, health record

and health database. After combined these terms, we

construct our search query: ("federated" AND

(health record OR health database)).

Searching by article title, abstract and keywords,

it was returned 42 articles on Scopus, 17 on ISI, at

PubMed 44 articles and 3 articles on Google. After

reading the title and abstract, we excluded all the

articles containing only one specific information

databases like genome, molecular. We only choose

articles whose content was related to federated

systems focused in patient’s health history. At the

end, we selected 16, 8, 12 and 3 articles respectively.

The final result was 14 articles. The search process

is represented in Figure 1.

Figure 1: Search flow diagram.

3 RESULTS

Through a literature search, we identified some

fundamental principles that must be respected

(Katehakis et al., 2001). Federated systems must

ensure interoperability using open standards; make

FEDERATED HEALTH REGISTERS - A Literature Review

159

sure of systems value through high quality services;

have modular architecture to allow developments,

maintenance and evolutions and so, enable

“monotonic” systems with an incremental evolution.

The creation of distributed clinical information,

maintaining autonomy, needs to be tested to avoid

conflicts that can occur in three levels (table 1)

(Román et al., 2006).

Table 1: Description of problems related to the different

levels of interoperability.

Levels Description of conflict

Semantic

Originate on different databases schemas that

were created independently. Interoperability is

only possible in previously common semantic

concepts.

Functional

Different components usually have different

functionalities and interfaces. A federated system

is based in a single identification and view. The

information must be exported functionality

according to the normalized interfaces.

Instance

Happens when the same person information’s

stored in different systems has to be merged. The

corresponding Personal ID handled by each

system must be found with the guarantee that it

refers the correct person and prevent a value

conflict.

3.1 Messaging versus Federation

Interoperability between systems can be made

through messages or federation.

3.1.1 Message-based

The message-based communication, in particular

based on HL7/DICOM, is considered as a

mechanism that facilitates the functional integration

of clinical information systems and administration,

institutional or regional level, resulting in the

automation of medical processes (Katehakis et al.,

2001).

This form of messaging is mainly used to share

only portions of the I-EHR and uses various

locations to store information, what results in

redundancy of information generated (which, in

many cases, can lead to inconsistencies) because it

focuses on episodes of care and referrals and

facilitates rapid entry of data, to cover a fairly large

number of end-user requirements.

3.1.2 Federated Approach

The federated approach is mainly used to promote

the virtual view of IEHR, without information

replication.

Any federated approach to an I-EHR

environment should be able to supply uniform

means of authentication, providing quick and

authorized access to personal health records; be

physician-generated and dispose patient record

information that is physically located in a different

clinical IT system (Katehakis et al., 2001).

3.2 Reference Architecture for the

Health Information Infrastructure

(HII)

The development of global information societies led

many countries to give high priority to create and

permit access to the I-EHR of a citizen. Therefore,

another priority is the creation of a health

information infrastructure (HII) to support the

provision of a variety of telematics and healthcare

services electronically (Tsiknakis et al., 2002).

A medical institution regional/national HII is

fundamentally about bringing timely health

information and aiding communication that brings

benefits for health decisors, their families, their

patients, and their communities. By this way,

individuals and public health professionals are HII

stakeholders and users, and the applications that

meet their respective needs are important

components of the infrastructure (Tsiknakis et al.,

2002).

Taken as a whole, the HII draws upon principles,

best practices, partnerships and necessary laws, but

is based on the use of standards, systems,

applications, and technologies that support

personalized healthcare services through the

effective information integration of networked

information sources (Tsiknakis et al., 2002).

The system’s architecture is a formal description

of an IT system, organized in a way that supports

reasoning about the structural properties of the

system. It defines the components that make up the

overall information system, and provides a plan to

implement the overall system. Usually is represented

by means of an architecture model. The Reference

Model Open Distributed Processing (RM-ODP) is

an architecture model used actively by industry in

the domain of healthcare that sets a standard of

reference for an open distributed processing

(Tsiknakis et al., 2002).

The purpose, therefore, of an architecture

regarding the technical aspects for developing a HII

is to provide and enable interoperability; modularity;

migration; stability; maintenance and cost-

effectiveness.

HEALTHINF 2012 - International Conference on Health Informatics

160

3.2.1 Reference Model for Open Distributed

Processing (RM-ODP)

RM-ODP is a reference model in computer science,

which provides a coordinated framework for the

standardization of open distributed processing

(ODP). It supports distribution, interworking,

platform and technology independence and

portability, together with an enterprise architecture

framework for the specification of ODP systems.

The RM-ODP view model provides five generic and

complementary viewpoints on the system and its

environment (Tsiknakis et al., 2002).

3.2.2 Components of I-EHR

Components of I-EHR can be distinguished in

generic and specific (table 2 and 3) (Katehakis et al.,

2001); (Tsiknakis et al., 2002).

Table 2: Generic components.

Generic

components

Description

Patient

Identification (PID)

Unique association of distributed

patient record and correlating their

IDs across different clinical

information systems.

Auditing (AUD)

Record all performed interactions

between middleware services and

final-user applications.

Authentication

(AUT)

Control access.

Encryption (ENC)

Secure communications of personal

information on Virtual Private

Network (VPN) as well on the

Internet;

Resource Location

Identify availability of related

resources

Terminology (TER)

Interpret and translate terms between

different coding schemes,

terminologies and internal semantics.

User profiles (UPR)

Maintain personalized settings and

preferences

Table 3: Specific components.

Specific components Description

Indexing Service

(IS)

Primary Health

Information Access

Services

Direct access of primary healthcare

clinical systems where the complete

and original clinical information is

kept.

Update Broker

Propagation to the IS of all

modifications pertaining to clinical

information

Components like AUT, TER and REL in

conjunction with others services can be used in

implementation of structures for telemedicine, home

care, clinical messaging, etc. (Katehakis et al.,

2001).

3.3 Standardization and Projects

A large variability of institutions like IEEE (Institute

for Electrical and Electronics Engineers), WHO

(World Health Organization), ANSI (American

National Standards Institute) and ASTM

(Association for Standards and Technology

Management) are working on the production of

standards in health domain. The results were the

creation of standards like HL7 (Health Level Seven),

CEN TC251 (European Standards for Health

Informatics), ISO TC215 (International SDO for

Health Informatics), GEHR (OpenEHR) and

DICOM (Diagnostic Imaging Communications).

“The diversity of standards difficult

interoperability. Data integration from

heterogeneous sources is difficult because

information systems differ in their functionality,

terminology, semantic, interface and internal data

representation” (California Healthcare Foundation.

Clinical Data Standards Explained. November

2004).

Some European projects of federated systems are

using some standards referred before like Synapses,

a pan-European project funded under the EU Health

Telematics Programme (Grimson et al., 1998);

(Tsiknakis et al., 2002); (Bisbal et al., 2003), that is

using GEHR, CENT/TC251 (Toussaint et al., 1997).

HYGEIAnet (Katehakis et al., 2001); (Katehakis et

al., 2001) in Crete, use OMG COAS (Object

Management Group -Clinical Observation Access

Service) (Tsiknakis et al., 2002). EHR in Portugal is

using OpenEHR, HL7, DICOM, CEN/TC251 –

EN13606. Medis (Sucurovic) is a project in Serbia

and Montenegro, that develops a prototype secure

national healthcare information system; and IHE

XDS (Dogac et al., 2007) is an idea to store the

healthcare document in an ebXML

registry/repository architecture to facilitate sharing.

3.3.1 Portuguese Project

The Portuguese Electronic Health Record

(Portuguese Health Ministry, 2009) by 2015 will

incorporate issues of consolidation in Europe and it

is an example of a federated register. It’s estimated

that by 2012 it will be already in operation on a set

of public and private entities. The solution provided

by that date, will then be gradually expanded,

functional and technologically, to all healthcare

providers.

FEDERATED HEALTH REGISTERS - A Literature Review

161

This model fits into a class that may be called

"federated model" because considers a distributed

form of delivery or data repositories, both physical

and logical resources that will give body to the

whole system and considers a form of management

that provides distinct areas of authority for different

sets of information and resources, cooperating,

according to commonly accepted rules, which

compliance is subject to verification and monitoring

of a central authority.

User's Health relevant data to the I-EHR are

stored in local repositories in the entities where they

are produced, at various healthcare consultations and

occurrences;

The transaction of the identified data is made by

Push form, e.g. the different entities systems push

data to the central EHR.

Model of distributed architecture based on three

levels of aggregation and information delivery:

Level 1 (Common Core): where the information

resides (central data repository) that will be available

centrally to all users, Citizens and Healthcare

Providers, including:

- A Minimum Set of Data (Summary);

- An "index" of the clinic history of the Citizen,

which will contain links to level 2 sites, where the

information related to various episodes of care will

be stored.

Level 2 (Sharing Level): where transitional

information will be provided, standardized,

produced by a particular entity as a result of

caregiving (data conceptually divided by domains of

responsibility associated with the various entities

that at this level would provide in a standardized

manner, the relevant information).

Level 3 (Specific Level): internal level and

reserved for a particular institution, where detailed

information produced within that institution will be,

residing in their own specific systems (data stored in

the repositories of their "clinical process" place

under the jurisdiction of each entity).

Ofélia, in Portugal, is a project to manage

federated identities and authorized mechanisms.

3.4 Security of I-EHR

Healthcare is a security sensitive domain with

confidential information.

Non authorized access to medical information

violates medical confidentiality or cause alterations

of medical data that may put patient’s health at risk.

Certification of the medical identity must be

considered essential for the final granting

authentication of medical doctors and is crucial for

any federated system.

This question may be solved using some

mechanisms of protection such as digital signatures

to protect validity, authenticity and integrity of

medical information as well as non-repudiation; use

of cryptography assures confidentiality and recipient

identification. To prevent access violations a reliable

auditing mechanism has to be employed (Tsiknakis

et al., 2002).

Minimal requests to respect in a secure domain

are described (Portuguese Health Ministry, 2009):

1) Confidentiality: Must allow definition of

sensibility levels and the classification of

information in sensibility levels; ensure the

confidentiality of the information in every step of his

life cycle; need a strong encryptions mechanisms

and resent in every versions of the information.

2) Integrity: Support the existence of versions to

save and modified information.

3) Availability: Should be able the application of

physical and logical mechanisms that ensure high

level of availability when occur need of access and

utilization.

4) Identity: Require mechanisms that ensure unique

and persistently the verification of user’s identity;

verification must be ensure in a integrated way with

national identity registration; Support electronic

signature associated to each user; allows control

about user’s identity and ensure permanently

compliance of the access control laws defined;

violation and access mechanism in emergency

situation.

5) Access Control: Must allow a definition of

utilization profile associated to users profession and

definition of access control by each user or profile.

6) Auditability: Register, monitoring and auditing

access activity must exist in a integrate way and all

operations must be auditable and registered on

system.

4 DISCUSSION

To develop infrastructures as the Federated Health

(Tsiknakis et al., 2002). The first one is a

definition/adoption of federated schemas that are

capable of supporting and providing effective

solutions to immediate needs without imposing

significant constraints in dealing with the issue of

incorporating new systems in the federation. So, the

establishment of the required consent from all

HEALTHINF 2012 - International Conference on Health Informatics

162

organizational units enable exporting and mapping

of their local schema. This process involves concept

mapping, the implementation of the corresponding

data extraction gateways, and the registration of the

new feeder systems into the federation's resource

directory. This enables clinical information systems

to push information to the middle layer of the health

information infrastructure.

The use of standardized interfaces for accessing

clinical information, either directly by the end user

or through the set of components residing at the

middle level of architecture's, managing the required

minimum data sets, allows the appropriate Human

Computer Interaction environment to support easy

and efficient access to the EHR data, as well as

indexing.

The implementation of the required mechanisms

enable information consistency and guarantee the

required Quality of Service and ensure an adequate

security system, with the consent management being

part of the overall security policy.

5 CONCLUSIONS

The future challenge for researchers and system

developers (Tsiknakis et al., 2002) is to provide a

new organizational framework that possibilities

integrate a diversity of heterogeneous resources;

increase the availability of previously inaccessible

information and address the demanding information

processing requirements of modern medical

applications.

REFERENCES

Bisbal, J., J. Grimson, et al., (2003). "From passive to

active electronic healthcare records." Methods of

Information in Medicine 42(5): 535-543.

Dogac, A., G. B. Laleci, et al., (2007). "Enhancing IHE

XDS for Federated Clinical Affinity Domain Support."

Information Technology in Biomedicine, IEEE

Transactions on 11(2): 213-221.

Grimson, W., D. Berry, et al. (1998). "Federated

healthcare record server--the Synapses paradigm."

International Journal of Medical Informatics 52(1-3):

3-27.

Katehakis, D. G., S. Kostomanolakis, et al., (2001). An

open, component-based information infrastructure to

support integrated regional healthcare networks.

Medinfo 2001: Proceedings of the 10th World

Congress on Medical Informatics, Pts 1 and 2. V. L.

Patel, R. Rogers and R. Haux. 84: 18-22.

Katehakis, D. G., S. Sfakianakis, et al., (2001).

Fundamental components for the realization of a

federated Integrated Electronic Health Record

environment. Engineering in Medicine and Biology

Society, 2001. Proceedings of the 23rd Annual

International Conference of the IEEE.

Portuguese Health Ministry, Health Systems Management

Center, (2009). EHR - State of art

Portuguese Health Ministry, Health Systems Management

Center (2009). Guidelines for Functional and

Technical System of EHR.

Muilu, J., L. Peltonen, et al., (2007). "The federated

database--a basis for biobank-based post-genome

studies, integrating phenome and genome data from

600,000 twin pairs in Europe." Eur J Hum Genet

15(7): 718-723.

Román, I., L. M. Roa, et al., (2006). "Demographic

management in a federated healthcare environment."

International Journal of Medical Informatics 75(9):

671-682.

Sucurovic, S. "Implementing security in a distributed web-

based EHCR." International Journal of Medical

Informatics 76(5-6): 491-496.

Toussaint, P. J., M. Kalshoven, et al., (1997). "Supporting

shared care for diabetes patients. The synapses

solution." Proc AMIA Annu Fall Symp: 393-397.

Tsiknakis, M., D. G. Katehakis, et al., (2002). "An open,

component-based information infrastructure for

integrated health information networks." International

Journal of Medical Informatics 68(1-3): 3-26.

FEDERATED HEALTH REGISTERS - A Literature Review

163